---
format: "aidr-story-markdown/v1"
id: "576ded508c0bb4010aee77b7ad6937092b8cb3bb40b0b89b851fa919bbf18228"
canonical_url: "https://aidr.today/576ded50?lang=en"
title: "AI is now capable of developing its own inference hardware"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-06T16:23:25.000Z"
category: "Open Source"
topics: ["inference","open-source"]
source_urls: ["https://github.com/FeSens/openTPU","https://news.ycombinator.com/item?id=49980715"]
summary: "An open-source AI accelerator developed by AI: RTL includes RTL, ISA, simulator, compiler and profiler in one repository and runs Qwen3, LFM2.5 and Qwen3.5 on a Kintex-7 PCIe card."
---

# AI is now capable of developing its own inference hardware

> [Open the canonical story](<https://aidr.today/576ded50?lang=en>)

**Published:** 2026-10-06T16:23:25.000Z
**Category:** Open Source
**Topics:** inference, open\-source

## Summary

An open\-source AI accelerator developed by AI: RTL includes RTL, ISA, simulator, compiler and profiler in one repository and runs Qwen3, LFM2\.5 and Qwen3\.5 on a Kintex\-7 PCIe card\.

## Sources

- [Story source](<https://github.com/FeSens/openTPU>)
- [Discussion](<https://news.ycombinator.com/item?id=49980715>)